Root Mean Squared Error (RMSE)

Calculates the root mean square of the model by taking the square root of mean of the sum of squares between the truth, \(y\), and the predicted, \(\hat{y}\) at each observation \(i\).

Usage

rmse(y, yhat)

Arguments

y

A vector of the true \(y\) values

yhat

A vector of predicted \(\hat{y}\) values.

Value

The RMSE in numeric form

Details

The formula for RMSE is: $$\sqrt {\frac{1}{n}\sum\limits_{i = 1}^n {{{\left( {{y_i} - {{\hat y}_i}} \right)}^2}} } $$

Examples

# Set seed for reproducibility
set.seed(100)

# Generate data
n = 1e2

y = rnorm(n)
yhat = rnorm(n, 0.5)

# Compute
o = mse(y, yhat)